Lyrl's recent activity
-
Comment on Our AI midwife in ~health
-
Comment on Our AI midwife in ~health
Lyrl Link ParentI am high risk for breast cancer due to radiation treatment for lymphoma, and talked with my oncologist and gynecologist about regular breast MRI. Neither one recommended it because the contrast...maybe we should just have MRIs as part of standard reproductive-related health exams...?
I am high risk for breast cancer due to radiation treatment for lymphoma, and talked with my oncologist and gynecologist about regular breast MRI. Neither one recommended it because the contrast agent they inject to get clear MRI pictures accumulates in fatty tissues, including the brain. Occasional use seems fine, but recurring use seems likely to be risky.
-
Comment on Trolley problem, again in ~humanities
Lyrl Link ParentVoting in a two-party system seems to generate quite similar arguments. Even just the choice to vote - not going any further with volunteering or donations or anything - people will object to...Voting in a two-party system seems to generate quite similar arguments. Even just the choice to vote - not going any further with volunteering or donations or anything - people will object to voting at all because they slightly dislike one choice and vehemently dislike the other. They commonly express the "pulling a (voting) lever is making me responsible for the choice, skipping voting leaves me outside the moral responsibility" language that is generally part of trolley problem discussions.
-
Comment on AI hallucination of Chinese nuclear components almost led to US military attack in ~society
Lyrl Link ParentI think those are all the points the people who leaked this incident to CNN were getting at. Optimally the problem would be recognized and worked on internally, but failing that trying to rally...I think those are all the points the people who leaked this incident to CNN were getting at. Optimally the problem would be recognized and worked on internally, but failing that trying to rally public pressure seems reasonable.
Quotes from the CNN sources I read as asking for attention to the need for AI-specific system improvements.
The analyst then used AI again to package the findings into a standard intelligence report — the kind that is trusted by military officials — and disseminated it.
“AI in targeting is definitely something that is ramping up and there is no real guidance for how having a human in the loop will prevent civilian casualties or fratricide,” another source familiar with the military’s current policies said.
The administration has made huge errors, as you detailed, but even if Hegseth et al are all removed from influence (may it be so, and quickly), the underlying systems problem still needs to be figured out. It bothers me to see responses that decry the leadership and completely ignore the challenges of figuring out new systems. And then I overcompensate by replying about the systems problems and don't mention the leadership failure.
-
Comment on AI hallucination of Chinese nuclear components almost led to US military attack in ~society
Lyrl Link ParentAgree the focus needs to be on the system and not the AI per se, but the AI introduces novel challenges there aren't yet good system examples to model off of. When AI fails, it does so in...Agree the focus needs to be on the system and not the AI per se, but the AI introduces novel challenges there aren't yet good system examples to model off of. When AI fails, it does so in different ways than a human would. This makes everyone's instincts developed from their own training, and any training they did of others, worse than useless. Focusing on the AI-human interaction to understand it is, I think, a reasonable first step to trying to figure out what system elements need to be set up.
-
Comment on AI hallucination of Chinese nuclear components almost led to US military attack in ~society
Lyrl Link ParentHumans react to the systems they are working within. Trying to figure out how to make systems that involve AI have characteristics that make humans more likely to use independent verification of...Humans react to the systems they are working within. Trying to figure out how to make systems that involve AI have characteristics that make humans more likely to use independent verification of key points, both the analysts originating intelligence reports and the officials deciding how to react to these reports, is IMO more important than blaming the analysts and officials.
-
Comment on The doomsday cult inside OpenAI in ~tech
Lyrl Link ParentA big part of the danger, as I understand it, is exposing and making it relatively easy to exploit software security vulnerabilities at rate not anticipated by standard security practices of the...A big part of the danger, as I understand it, is exposing and making it relatively easy to exploit software security vulnerabilities at rate not anticipated by standard security practices of the last several years or decades. Standards can adapt, vulnerabilities can be patched, and if you as the "good guy" give friendly actors tools and information that helps them make that transition, then "your side" will be at an advantage. Compared to stopping building of the thing, the "bad guys" get the tools first, and "your side" suffers. I don't think unilateral stand down would make sense to the people involved in AI development, even those whose fears of the technology are sincere.
-
Comment on Off the Norwegian island of Fedje lies a German World War II submarine believed to contain a huge load of mercury – why salvaging it has become a race against time in ~humanities.history
Lyrl Link ParentA company I worked for once had a contractor freeze a 3-inch diameter water pipe to enable a shutoff valve to be installed. (Had been forgotten in initial construction and not noticed until we...A company I worked for once had a contractor freeze a 3-inch diameter water pipe to enable a shutoff valve to be installed. (Had been forgotten in initial construction and not noticed until we needed it a decade later.) Several companies no-quoted due to the pipe diameter - they only froze up to 2-inch pipes - and when our job was done, the freezing didn't go all the way through: it just slowed down the fresh water flow enough the valve installation was messy but doable. That experience makes me skeptical that freezing saltwater at pressure while surrounded by more water would be reliable.
Part of the Fukushima containment was freezing surrounding soil, validating ice as a containment strategy, I just don't know of a way to be confident in an underwater freezing of saltwater. Especially when, like asbestos, messing with the mercury unsuccessfully would be way worse than just leaving it alone.
-
Comment on Gemini AI hacked three companies in a testing breakout, Google says in ~tech
Lyrl Link ParentIt's widely believed that aviation safety is better because accident investigations are focused on preventing future harms and not on assigning blame. Seeking prosecutions is actively harmful to...It's widely believed that aviation safety is better because accident investigations are focused on preventing future harms and not on assigning blame. Seeking prosecutions is actively harmful to making the system safer. https://www.realclearinvestigations.com/articles/2024/02/09/investigative_issues_why_youve_never_been_in_a_plane_crash_1010766.html
I believe AI incidents are similarly complex and multi-factor as aviation accidents, and would feel less safe if prosecution activity overshadowed the process of figuring out how to prevent future incidents.
-
Comment on Weekly US politics news and updates thread - week of September 14 in ~society
Lyrl Link ParentIt's an interesting open question how much of China's AI progress is their own innovation, and how much is distillation of the latest western models. If it's mostly distillation, slowing down...It's an interesting open question how much of China's AI progress is their own innovation, and how much is distillation of the latest western models. If it's mostly distillation, slowing down American companies would also slow down Chinese ones. Kind of high stakes to be taking a guess at, but if our intelligence agencies were led better maybe we wouldn't have to be guessing.
-
Comment on What are your "how on earth did I not learn this sooner" tech moments? in ~tech
Lyrl Link ParentAlt plus click and drag allows column highlighting, so you can select some text from multiple lines without having to select an entire line.Alt plus click and drag allows column highlighting, so you can select some text from multiple lines without having to select an entire line.
-
Comment on We surveyed 6,000 Americans. Here is the winning message for Democrats - Opinion, Dustin Guastella and Erica Etelson in ~society
Lyrl Link ParentWhen I have come across leftist threads on social media complaining about not gaining traction, I would classify the philosophy as what this group described as the "divide-and-conquer playbook". I...When I have come across leftist threads on social media complaining about not gaining traction, I would classify the philosophy as what this group described as the "divide-and-conquer playbook".
I think the kitchen table option being promoted as feasible now would not have polled so well a couple of years ago. That was before tariff-driven inflation, before Iran-war-driven inflation, and many people who wanted to believe that Trump, as a strongman, could just tell issues to be fixed and life would get better (but not in his first term because Deep State) had no counterfactuals to contend with.
-
Comment on Older adults who typically had longer intervals between meals accumulated chronic diseases more rapidly over time than those with shorter intervals, according to study from Karolinska Institutet in ~health
Lyrl LinkFull study article: https://onlinelibrary.wiley.com/doi/10.1111/joim.70150 That half of participants reported longest fast of 12-13.5 hours is a relatively tight grouping around what I would......people whose longest interval between meals during a typical day was 14 to 24 hours accumulated chronic diseases more rapidly than those whose longest interval was 6 to 11.5 hours.
Full study article: https://onlinelibrary.wiley.com/doi/10.1111/joim.70150
That half of participants reported longest fast of 12-13.5 hours is a relatively tight grouping around what I would consider a normal dinner to breakfast interval, which feels validating to my concept of mealtimes.
The focus on longest interval wasn't what I expected from the headline. Also, 14 hours from dinner to breakfast seems kind of short to be risky. I wonder if that low end of the "long interval" quartile, something like 6pm dinner and 8am breakfast, is actually risky, or if the grouping into quartiles with people only eating once a day skewed the statistics.
Also, longest interval of 6 hours, who does that? I guess bedtime snack at midnight and breakfast at 6am?
-
Comment on Invite request spam problem - vet through other social media? in ~tildes
Lyrl Link ParentNeat to learn, thanks for the pointer! It looks like @cfabbro is still holding down the fort over there. I am thinking of starting something similar on lemmy, and will look over that sub reddit...Neat to learn, thanks for the pointer! It looks like @cfabbro is still holding down the fort over there. I am thinking of starting something similar on lemmy, and will look over that sub reddit setup for ideas.
-
Comment on Invite request spam problem - vet through other social media? in ~tildes
Lyrl Link Parent@Deimos - I have seen speculation in a variety of threads on current invite policy implied by invites not appearing without request anymore, neither for new users or for people who use up their...@Deimos - I have seen speculation in a variety of threads on current invite policy implied by invites not appearing without request anymore, neither for new users or for people who use up their most recent set of 10. Everything from a change to on-request-only invite generation just to reduce your maintenance overhead (i.e. no policy change), to new user probation policy, to intentional growth restriction. Are you willing to share an official statement on availability of invites?
-
Invite request spam problem - vet through other social media?
For most of the history of Tildes, if someone came across the site and decided they wanted to join, but didn't know anyone already on, they could email Deimos with an invite request. However, with...
For most of the history of Tildes, if someone came across the site and decided they wanted to join, but didn't know anyone already on, they could email Deimos with an invite request. However, with the rise of LLMs, the noise-to-signal ratio has become so high this is no longer a way to join Tildes - explanation by Deimos: https://tildes.net/~tildes/1vl3/is_there_any_interest_in_building_a_tildes_successor#comment-if6r
I've read the recent discussions on Tildes being maintained as is and not actively grown, but I think most of us believe the community is up for many more years of enjoying each other's company, and would like to pick up a few more members along the way if they seek us out. That just requesting to join isn't possible anymore makes me a little sad.
I had the thought that larger sites are able to invest more resources into spam account filtering, many of us have active accounts on other social media sites (or other types of sites with community interactions), and I speculate that almost everyone who finds tildes and is interested in joining has already been active in some other online community. Some of us Tilderinos could post here our handles in other communities, and offer to consider sharing one of our invite requests to Tildes if DM'd in our other community. We'd have to reject requests from new accounts, and use some judgement on established ones, but I am thinking having people go through the filtering process those other communities use would reduce the spam requests enough to make it workable.
What do you guys think?
21 votes -
Comment on <deleted topic> in ~food
Lyrl Link ParentThere is some amount of genetic ability to taste, also. My mom gardenened through much of my childhood (and still does) and I would consider her good at it, but I never got the enjoyment others...There is some amount of genetic ability to taste, also. My mom gardenened through much of my childhood (and still does) and I would consider her good at it, but I never got the enjoyment others seem to out of home grown tomatoes. I once went to the Seed Saver's farm annual Tomato Tasting Festival - presumably they are good at growing tomatoes since they farm for a living - and the tomatoes were fine, I guess.
I believe others when they describe the awesomeness they experience eating homegrown tomatoes. But I am somehow not wired to experience that. I enjoy growing edible plants, but tomatoes aren't worth my time to grow.
-
Comment on Is there any interest in building a Tildes successor? in ~tildes
Lyrl Link Parent@Deimos - is non-response to emailed requests for invites official policy? If so, could the contact page https://docs.tildes.net/contact be updated? It seems unfriendly to direct people to send an...@Deimos - is non-response to emailed requests for invites official policy? If so, could the contact page https://docs.tildes.net/contact be updated? It seems unfriendly to direct people to send an email and claim an invite will be issued if that is no longer true.
-
Comment on Greenland issues ‘strong warning’ as Donald Trump-linked oil firm prepares to drill in ~society
Lyrl Link ParentThey might be trying to delay confrontation until a new administration is in place in two and a half years. That's a long time to slow walk things, but if they can pull it off would be the path...They might be trying to delay confrontation until a new administration is in place in two and a half years. That's a long time to slow walk things, but if they can pull it off would be the path least likely to have militia-like escalation.
-
Comment on US general warns he lacks sufficient forces to protect Israel amid Iran war in ~society
Lyrl Link ParentDeferred maintenance on critical navy vessels is a horrible ticking time bomb. It was a significant contributing factor to the accidents that killed seventeen sailors ten years ago...Deferred maintenance on critical navy vessels is a horrible ticking time bomb. It was a significant contributing factor to the accidents that killed seventeen sailors ten years ago (https://features.propublica.org/navy-accidents/us-navy-crashes-japan-cause-mccain/). That lesson being thrown out the window, but this time in a war zone is incredibly painful.
They mention in the article laparascopy surgery of the abdomen visualizing the outside of the uterus, hysteroscopy, and ultrasound of the uterus. The author doesn't go into very much detail about the rest of their 6-year treatment, but it seems likely other imaging of the uterus was also done. It wasn't a failure to do any imaging, it was a failure to do yet another imaging after many imagings that hadn't found anything. Or failure to pick up on the potentially suspicious spot on the ultrasound that the AI noticed and based its recommendation of an MRI follow up on.